Roles of N-Vacancies over Porous g-C3N4 Microtubes during

Feb 26, 2019 - Roles of N-Vacancies over Porous g-C3N4 Microtubes during Photocatalytic NOx Removal. Zhenyu Wang†‡§ , Yu Huang*†‡ , Meijuan C...
0 downloads 0 Views 2MB Size
Subscriber access provided by Macquarie University

Energy, Environmental, and Catalysis Applications

The Roles of N-vacancies over Porous g-C3N4 Microtubes during Photocatalytic NOx Removal Zhenyu Wang, Yu Huang, Meijuan Chen, Xianjin Shi, Yufei Zhang, Jun-ji Cao, WingKei Ho, and Shun Cheng Lee ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.8b21987 • Publication Date (Web): 26 Feb 2019 Downloaded from http://pubs.acs.org on February 27, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 51 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

The Roles of N-vacancies over Porous g-C3N4 Microtubes during Photocatalytic NOx Removal Zhenyu Wang a,b,c,

a

a,b,c,

Yu Huang

a,b*,

Meijuan Chen c, Xianjin Shi a,b, Yufei Zhang a,b, Junji Cao

Wingkei Ho d and Shun Cheng Lee e1

Key Lab of Aerosol Chemistry & Physics, State Key Lab of Loess and Quaternary Geology

(SKLLQG), Institute of Earth Environment, Chinese Academy of Sciences (CAS), Xi’an 710061, P. R. China b

CAS Center for Excellence in Quaternary Science and Global Change, Xi’an 710061, P. R.

China c

School of Human Settlements and Civil Engineering, Xi’an Jiaotong University, Xi’an

710049, P. R. China d

Department of Science and Environmental Studies, The Education University of Hong

Kong , Hong Kong, P. R. China eDepartment

of Civil and Environmental Engineering, The Hong Kong Polytechnic

University, Hong Kong, P. R. China Submitted to: ACS Applied Materials & Interfaces

KEYWORDS: N-vacancy; tubular g-C3N4; porosity; photocatalytic NOx removal *Corresponding author: Prof. Yu Huang, E-mail address: [email protected] Tel: 86-29-6233 6261 1 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 51

ABSTRACT

The development of catalysts that effectively activate target pollutants and promote their complete conversion is an admirable objective in the environmental photocatalysis field. In this work, graphitic carbon nitride (g-C3N4) microtubes with tunable N-vacancy concentrations were controllably fabricated using an in-situ soft-chemical method. The morphological evolution of g-C3N4, from the bulk to the porous tubular architecture, is discussed in detail with the aid of time-resolved hydrothermal experiments. We found that the NO removal ratio and apparent reaction rate constant of the g-C3N4 microtubes were 1.8- and 2.6-times higher than those of pristine g-C3N4, respectively. By combining detailed experimental characterization and DFT calculations, the effects of N-vacancies in the g-C3N4 microtubes on O2- and NO-adsorption activation, electron capture, and electronic structure, were systematically investigated. These results demonstrate that surface N-vacancies act as specific sites for the adsorption-activation of reactants and photo-induced electron capture, while enhancing the light-absorbing capability of g-C3N4. Moreover, the porous wall structures of the as-prepared g-C3N4 microtubes facilitate the diffusion of reactants, and their tubular architectures favor the oriented transfer of charge carriers. The intermediates formed during photocatalytic NO-removal processes were identified by in-situ diffuse reflectance infrared Fourier transform spectroscopy, and different reaction pathways over pristine and N-deficient g-C3N4 are proposed. This study provides a feasible strategy for air-pollution control

over

g-C3N4

by

introducing

N-vacancy

and

porous-tubular

architecture

simultaneously.

2 ACS Paragon Plus Environment

Page 3 of 51 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

INTRODUCTION Nitric oxides (NOx) are among the most important ozone- and haze-forming precursors.1-2 Transforming and using solar energy to trigger the photocatalytic removal NOx under ambient conditions has attracted tremendous attention.3 Photogenerated active species can oxidize gas-phase NOx to particulate-phase nitrate, which purifies the air.4-9 Compared to traditional metallic-oxide catalysts (e.g., TiO2 and SrTiO3), graphitic carbon nitride (g-C3N4) with fascinating 2D graphite-like layered structure has received significant attention as a nonmetallic and visible-light-responsive photocatalyst for NOx-degradation applications.10-17 Previous studies have also demonstrated that superoxide (•O2−) and photogenerated holes (h+) are the dominant active species employed by g-C3N4 to remove NOx.11, 18-19 Unfortunately, pristine g-C3N4, with its irregular bulk morphology, removes NOx inefficiently. On the one hand, pristine g-C3N4, with an intact tri-s-triazine structure, shows limited ability to adsorption-activate target reactants (O2 and NO) and absorbs visible light poorly due to the lack of redundant coordination bonds and defect states in its electronic structure, respectively.20-21 On the other hand, photogenerated charge carriers diffuse randomly in the bulk g-C3N4 due to the lack of an engineered morphology and electron-capture sites, which promotes the recombination of electron-hole pairs.11, 22-23 With the aim of remedying these deficiencies, constructing adsorption-activation sites, modifying the electronic structure, and engineering the morphology of g-C3N4 are expected to be effective strategies. Intrinsic anionic vacancies are important for heterogeneous reactions because they improve the light-absorbing abilities of materials while acting as specific sites that adsorption-activate reactant molecules and capture electrons.24-25 The positive roles of oxygen vacancies 3 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 51

(O-vacancies), typical anionic vacancies in oxides, in O2 activation, electron capture, and the efficient use of incident light have been actively pursued.26-29 Hence, inspired by findings related to O-vacancy materials, nitrogen vacancies (N-vacancies, typical anionic vacancies in nitrides) in g-C3N4 have the potential to effectively activate target pollutants, capture photogenerated charge carriers, and engineer the band structure; however challenges still remain. Current approaches for introducing N-vacancies into g-C3N4 involves calcining processors under different atmospheric conditions, or calcining at different temperatures in inert gases;30-31 however, these approaches cannot engineering the morphology of the material. In addition, the roles of N-vacancies in electronic-structure engineering, and in accelerating photogenerated charge-carrier separation, have been investigated,32-33 but the effect of the N-vacancy spatial distribution on electron-separation efficiency was ignored. Notably, the adsorption-activation behavior of target reactants (O2 and NO) on N-deficient surfaces has not been investigated. Recently, the morphology engineering of irregular architectures into one-dimensional (1D) tubular structures has strengthened interest in the oriented transfer of charge carriers.34 One of the pioneering studies in this area reported that the 1D tubular structure of TiO2 acted as a highway for the oriented transfer of electrons and exhibited superior charge-carrier separation and transport capabilities as a consequence.35 Therefore, constructing 1D tubular g-C3N4 structures that constrain random photogenerated charge-carrier diffusion is feasible. In recent years, calcining rod-like melamine·cyanuric acid (M·CA) complex precursors to produce tubular g-C3N4 has been regarded to be a simple and reliable method because it avoids tedious steps and the introduction of impurities. The rod-like precursor of supramolecular M·CA is 4 ACS Paragon Plus Environment

Page 5 of 51 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

fabricated by mixing equal amounts of melamine and cyanuric acid followed by heating at 180 °C.36-37 Melamine is well known spontaneously transform into cyanuric acid under suitable conditions. Expensive cyanuric acid is avoided if monomeric melamine is used to fabricate supramolecular M·CA in an in-situ self-conversion process. To date, the easy transfer of electrons in the longitudinal direction of tubular g-C3N4 has been widely confirmed; however, charge-carrier separation efficiency remains unsatisfactory unless g-C3N4 is doped with foreign atoms or by constructing isotypical heterostructures.38-40 In addition, the enhancement of reactant diffusion is another important function of morphology engineering, and the formation of porous structures is an effective strategy to achieve this. In our previous work, we fabricated honeycomb-like g-C3N4 that exhibited improved NOx-diffusion abilities compared to pristine g-C3N4.41 Therefore, the preparation of porous g-C3N4 with a 1D tubular architecture is highly desirable and is expected to accelerate the simultaneous oriented transfer of photo-induced e−-h+ pairs, as well as reactant diffusion. In this work, a series of g-C3N4 microtubes with tunable N-vacancy concentrations and porous wall structures were synthesized by an in-situ soft-chemical method. The novel synthesis involved calcining N-deficient rod-like precursors which were synthesized by the self-conversion of monomeric melamine for different hydrothermal-treatment times. The morphological evolution of the porous-tubular architecture is discussed in detail, aided by time-resolved hydrothermal experiments. The prepared porous g-C3N4 microtubes exhibited clearly improved photocatalytic activity for NOx degradation compared to bulk g-C3N4. The effects of N-vacancies on O2- and NO-adsorption activation, electron capture, and electronic structure, and the effect of the tubular structure on oriented electron transfer, were 5 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 51

systematically investigated through experimental and computational studies, which led to the proposal of a mechanism for the enhanced NO-removal activity. Furthermore, the reaction pathways for photocatalytic NO removal over pristine g-C3N4 and N-deficient g-C3N4 were identified by diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS), and a mechanism for photocatalytic NO removal is proposed. This study provides a feasible strategy for air-pollution control over g-C3N4 by introducing N-vacancy and porous-tubular architecture simultaneously.

MATERIALS AND METHODS Synthesis of porous g-C3N4 microtubes with various N-vacancy concentrations Synthetic materials for g-C3N4 and other related chemicals were bought from Shanghai Aladdin Bio-Chem Technology Co., Ltd. All chemicals were used without further purification. Firstly, 1 g of melamine was added in 70 mL of deionized water and then the suspension liquid was heated to 60 °C in a water bath kettle with vigorous stirring until dissolution. Then, the obtained solution was transferred into a 100-mL Teflon-lined autoclave and heated at 180 °C for 6, 7, 8, 12, or 16 h; the precursors prepared in this matter are designated: M-6, M-7, M-8, M-12 and M-16, respectively. The hydrothermal products were filtered and washed with deionized water and absolute ethanol for three times, and subsequently dried at 60 °C for 12 h. Finally, 1 g of M-8, M-12, and M-16 were separately calcined to 550 °C in a muffle furnace for 3 h under a flow of nitrogen (2.5 °C/min); the resulting materials are designated: CNT-8, CNT-12, and CNT-16, respectively. For comparison, pristine g-C3N4 (designated as CN) was synthesized by directly calcining 1 g of 6 ACS Paragon Plus Environment

Page 7 of 51 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

the precursor prepared by drying solution A at 60 °C.

Characterization X-ray diffraction (XRD) of the samples were conducted on a X'Pert PRO X-ray diffractometer using Cu Kα radiation (PANalytical, UK, 2θ scan rate = 0.05 /s, λ = 1.5406 Å, 40 kV, 40 mA). Elemental compositions of the samples were performed on a Elementar Vario EL instrument (Vario EL III, Germany detection limit is 0.015%, standard deviation ≤ 0.1% abs). Fourier-transform infrared spectroscopy (FT-IR) of all samples were recorded on a VERTEX 70 FT-IR spectrometer (Bruker, Germany). The surface elemental compositions and states of the samples were conducted on an ESCALAB 250 X-ray photoelectron spectrometer (Thermo Fisher, USA). All binding energies were calibrated against the C 1s peak (284.6 eV) which arises from surface-adventitious carbon. The morphologies of the samples were examined using a scanning electron microscopy (SEM, JEOL JSM-6490, Japan) and a transmission electron microscopy (TEM, JEOL JEM-2100HR CM-120, Japan), respectively. The BET surface areas of samples were recorded on an ASAP 2020 nitrogen-adsorption analyzer (Micromeritics, USA); the samples were degassed at 250 °C prior to analysis. The free radicals were detected by sample trapping with a 25 mM solution of 5,5’-dimethyl-1-pyrroline-N-oxide (DMPO) on a ER200-SRC electron spin resonance (ESR) spectrometer (Bruker, Germany) at low temperature (130 K); aqueous dispersions were used for DMPO-•OH, and alcohol dispersions for DMPO-•O2−. The UV-vis absorption spectra of the samples were recorded on a Varian Cary 100 Scan UV-visible spectrophotometer (Agilent, Australia). The intermediate and final products were detected 7 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 51

using ion chromatography on a Dionex-600 Ion Chromatograph (IC, Dionex Inc., Sunny-vale, CA, USA) fitted with an IonPac AS14A column. Photoluminescence (PL) spectra were acquired on an FLS980 fluorescence spectrophotometer (Edinburgh, UK).

Computational modeling Density functional theory (DFT) calculations using the Cambridge Sequential Total Energy Package (CASTEP) software package. The plane-wave basis sets were used to treat valence electrons, while norm-conserving pseudo-potentials were used to approximate the potential field.42-43 A model of the g-C3N4 crystal was built through cleavage of the (001) lattice plane, followed by the construction of a 10-Å vacuum slab crystal, as shown in Figure S1a. The g-C3N4-crystal model with a N-vacancy was constructed by removing a two-coordinated nitrogen atom (Figure S1b). The exchange-correlation Perdew–Burke–Ernzerh (PBE) functional was implemented to describe the exchange-correlation energy and electron interactions.44 The Brillouin zone was separately sampled at 4  4  1 Monkhorst-Pack k-points.45 A plane-wave cutoff was set to 380 eV for all the calculations. During geometry optimization, the convergence criterion was set to 1.0 e-6 eV/atom for energy, and 0.05 eV/Å for residual force. The electron-density difference (EDD) and Mulliken population analysis (MPA) were used to calculate charge-transfer quantity.46 The O2 and NO adsorption energies (ΔEads) for the model g-C3N4 crystal without and with the N-vacancy were calculated using eqs. 1 and 2: ∆Eads = EO2 + C3N4 - EO2 - EC3N4

(1),

∆Eads = ENO + C3N4 - ENO - EC3N4

(2), 8

ACS Paragon Plus Environment

Page 9 of 51 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

where ΔEads is the O2 or NO adsorption energy, EO2+C3N4 and ENO+C3N4 are the total energies of the model g-C3N4 crystal with molecular O2 and NO adsorbed on its surface, respectively, ΔEC3N4 is the total energy of the model pure g-C3N4 crystal, EO2 is the energy (-869.17 eV) of molecular O2 in the gas phase, and ENO is the energy (-705.07 eV) of molecular NO in the gas phase.

Photocatalytic activity experiments The photocatalytic activities of the as-prepared samples towards NOx removal were conducted in a continuous-flow-reactor system. 0.1 g of the as-prepared sample were coated onto a 10.0-cm-diameter vitreous culture dish and then placed in the 4.5 L (30 cm × 15 cm × 10 cm) of rectangular reactor. The initial NO gas was supplied from a compressed gas cylinder (48 ppm N2 balance) and diluted to 400 ppb by a zero-air generator (Model 1001, Sabio Instruments Inc., Georgetown, TX, USA). A 300-W commercial Xe arc lamp (Perfectlight, China) fitted with a UV-cutoff filter was placed vertically outside of the reactor and turned on following NO-adsorption-desorption equilibrium. The optical power density at the sample surface was calibrated to be 25.28 mW/cm2 using an optical power meter (Thorlabs PM100D, Newton, NJ, USA). The change of NOx-concentrations were recorded on-line on a Model 42c chemiluminescence NO analyzer (Thermo Environmental Instruments Inc., Franklin, MA, USA). The removal ratio (η) of NO was calculated as η (%) = (1−C/C0) × 100, where C and C0 are the concentrations of NO in the outlet and feed streams, respectively.

9 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 51

Photoelectrochemical measurements The

photoelectrochemical

experiments

were

evaluated

using

a

Parstat

4000

electrochemical workstation (USA) fitted with three-electrode cell configuration; platinum plate electrode is counter electrode and Ag/AgCl electrode is reference electrode; the as-prepared sample was coated on ITO substrate and used as working electrode. For the transient photocurrent measurement, the Na2SO3 solution (0.5 mol/L) was served as the electrolyte and the photocurrent-time curves were measured at 0.2 V vs. Ag/AgCl at ambient temperature under a periodic visible-light illumination (Xe lamp, 300 W, Perfectlight, China). For the electrochemical impedance spectroscopy (EIS) measurement, the working electrode was immersed in a 1 mmol/L K3Fe(CN)6 and K4Fe(CN)6 solution, and subjected 5 mV voltage amplitude at a frequency range of 0.1 Hz to 100 kHz.

In-situ DRIFTS for NO adsorption-oxidation The DRIFTS measurement was performed on a VERTEX 70 FT-IR spectrometer (Bruker, Germany) fitted with a Harrick in-situ diffuse-reflectance-cell and a high-sensitivity mercury cadmium telluride (MCT) detector. The as-prepared sample was preheated in helium at 250 oC

for 45 min to remove surface adsorbates. After cooling to room temperature, the

background measurement was conducted. Then, a gas mixture (40% NO and 60% O2) was fed into the reaction chamber with a total flow of 50 mL/min. After 20 min, the visible light was introduced into the reaction chamber and removed after 30 min later.

RESULTS AND DISCUSSION 10 ACS Paragon Plus Environment

Page 11 of 51 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Phase structures, chemical components, and N-vacancy analyses. The phase structures of CN, and the CNT-8, CNT-12, and CNT-16 series of samples were investigated by XRD. As shown in Figure 1a, the CN sample exhibits two typical diffraction peaks at 13.2° and 27.3°; the former is assigned to intralayer long-range-ordered tri-s-triazine units (100) and the latter is assigned to periodic stacking (002) of layers along the c-axis.13 The (100) peak gradually decreases in intensity in moving to CNT-8 and CNT-12, and is almost absent in the spectrum of CNT-16, which is suggestive of the progressive loss of the periodic structure of the tri-s-triazine framework. To further identify the structural defects in the bulk phases of CNT-8, 12, and 16, these materials were first subjected to elemental analysis. The N and C contents of the precursors and prepared g-C3N4 are listed in Table S1, with the calculated N/C atomic ratios summarized in Table S2. The N/C atomic ratio was observed to decrease from 2.057 for pristine precursor M, to 1.537 for M-16, while the N/C atomic ratio of the g-C3N4 samples were determined to be 1.521 for CN, and 1.504, 1.489, and 1.427 for CNT-8, CNT-12, and CNT-16, respectively. The decreasing N content is consistent with the introduction of N-vacancies that destroy the ordered in-plane structure, leading to the gradual disappearance of the (100) peak. Compared with the CN sample, there was 1.1, 2.1, and 6.2 at% less nitrogen in CNT-8, CNT-12, and CNT-16, respectively. In addition, compared with CN, the (002) peaks of the as-prepared CNT samples gradually weakened and broadened, which is suggestive of a reduced crystal size. In Figure 1b, the FT-IR spectrum of CN exhibits typical bands at 810 cm-1 for the breathing mode of the heptazine ring system. The broad 1200–1600 cm-1 correspond to the stretching vibrations of the C6N7 units.47 The broad 3000–3600 cm-1 regions (shaded in orange) correspond to the 11 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 51

stretching vibrations of N-H.48 The remaining CNT samples exhibit distinctly enhanced peaks in the orange shaded regions, which are suggestive of a gradual increase in the concentration of N-H groups, which results in the loss of the in-plane periodicity of the tri-s-triazine framework. The surface elemental compositions and states of the as-prepared samples were further investigated using XPS. Figure 1c reveals that CN exhibits three C 1s peaks at around 284.6 eV, 286.3, and 288.2 eV, which correspond to adventitious C-C, C-NH/NH2, and N-C=N bonds, respectively (Table S3).33, 49-50 N 1s peaks are observed at around 398.5, 400.0 and 401.0 eV, which are assigned to two-coordinated (N2c), three-coordinated (N3c) nitrogen atoms and NHx groups (Table S3), respectively.33, 49, 51 The N/C atomic ratios on the sample surfaces were calculated and reveal a decreasing trend, from 1.361 for CN, to 1.322 for CNT-8, 1.299 for CNT-12, and 1.244 for CNT-16 (Table S2), consistent with a gradual increase in the N-vacancy concentration. In addition, the N2c/N3c peak-area ratio was observed to decrease, from 4.92 for CN to 4.72, 4.46, and 4.37 for CNT-8, CNT-12, and CNT-16, respectively (Table S2), indicating that N2c atoms are preferentially lost. The preferential loss of N2c atoms over N3c atoms is attributed to their unsaturated coordinations; original charge balances no longer exists with N2c atoms missing. To illustrate the charge redistribution, the EDDs of g-C3N4 and N2c-deficient g-C3N4 were calculated by DFT, the results of which are shown in Figure 1d. The N atom at the N2c site obtains 0.39 electrons from its neighboring C atoms in pure g-C3N4 (Figure 1d, left). After removal of the N2c atom, the density of the electron cloud increased to 0.52 electrons due to the lack of an electron acceptor (Figure 1d, right). The oxygen concentrations of CN, CNT-8, CNT-12, CNT-16 12 ACS Paragon Plus Environment

Page 13 of 51 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

were analyzed and the results were listed in Table S4. From the results of Table S4 and high-resolution O 1s XPS (Figure S2), the low concentrations of oxygens (2.13 Atomic % for CN, 0.97 Atomic % for CNT-8, 1.62 Atomic % for CNT-12, 1.14 Atomic % for CNT-16) of all the samples were from the surface adsorbed water.33 The relationship between unpaired-electron variation and N-vacancy concentration were further probed by ESR spectroscopy. As shown in Figure 1e, CN exhibits an insignificant single-line ESR signal at a g-value near 2.0036, which suggests that there are few localized unpaired electrons present in pristine CN.21 However, increasing spin intensities were observed for the CNT samples, consistent with a gradual increase in the N-vacancy concentration in moving to CNT-8, CNT-12, and CNT-16.

13 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 1. (a) XRD patterns, and (b) FT-IR, (c) high-resolution C 1s and N 1s XPS, and (e) ESR spectra of CN, CNT-8, CNT-12, and CNT-16. (d) Calculated EDDs diagrams of g-C3N4 (left) and N2c-deficient g-C3N4 (right). Morphologies and formation mechanism of porous g-C3N4 microtubes The morphologies of representative samples were examined by SEM and TEM. As shown in Figures 2a–c, the SEM images of CNT-8, CNT-12, and CNT-16 exhibit similar open-ended tubular structures that are more than 10 μm long and about 3 μm wide, which is distinctly different from that of the CN obtain by calcining melamine directly (Figure S3a). 14 ACS Paragon Plus Environment

Page 14 of 51

Page 15 of 51 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

TEM further confirmed that hollow tubular structures were successfully produced. The thickness of the tube wall gradually decreased in moving to CNT-8 (Figure S3b), CNT-12 (Figure 2d), and CNT-16 (Figure S3c). A magnified image of the tube wall of CNT-12 (Figure 2e) shows a large number of 20–50-nm nanopores embedded in the lamellar tube wall. These porous structures are formed in the tube walls in association with gases released during pyrolysis condensation. These porous surfaces facilitate the diffusion of reactants during heterogeneous reactions.41, 52 The SEM and TEM results demonstrate that the in-situ soft-chemical method is an effective approach for the construction of the porous tubular structure of g-C3N4. The textural properties of these g-C3N4 samples were further investigated by acquiring their nitrogen adsorption-desorption isotherms and Barrett–Joyner–Halenda (BJH) pore-size distributions. The specific surface areas of CN, CNT-8, CNT-12, and CNT-16 were found to be 19.9, 28.2, 67.5, and 69.2 m2/g, respectively (Figure 2f). In addition, the 2–4-nm pore-size distribution of CN reflects the porosity within its nanoscale sheets, while the distribution observed at 20–50-nm is attributing to layer-stacked pores. The CNT samples clearly exhibit more-intense distribution peaks in these regions due to the effects of layer exfoliation and pore formation, respectively, as evidenced by TEM. Hence, compared to CN, the increased specific surface areas of the CNT samples are attributable to the introduction of porous tubular structures.

15 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 51

Figure 2. SEM images of (a) CNT-8, (b) CNT-12, and (c) CNT-16. (d) TEM image of CNT-12. (e) Enlarged TEM image of the tubular wall of CNT-12. (f) N2 adsorption-desorption isotherms of CN, CNT-8, CNT-12, and CNT-16, and the corresponding pore-size distribution curves. 16 ACS Paragon Plus Environment

Page 17 of 51 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

The morphological evolution of tubular g-C3N4, taking CNT-12 as an example, is displayed through a time-resolved hydrothermal experiment, as shown in Scheme 1. Simply, bulk pristine melamine exhibits an irregular appearance (Figure S4a, Scheme 1, I) and a monoclinic crystal structure (Figure S4d, e).53 Melamine retains its bulk structure after hydrothermally reacting at 180 °C for 6 h, but its (102) crystal faces are exposed for growth (Figure S4b, d, Scheme 1, II). At a hydrothermal-reaction time of 7 h, both bulk and rod-like crystal structures coexist in M-7 (Figure S4c, Scheme 1, III). These new rod-like structures are M·CA supramolecular structures (Figure S4d, g), suggesting that: (1) M was partly converted into CA in-situ, and (2) the remaining M aggregated with CA to form M·CA supramolecular structures through hydrogen bonding (N–H...O and N–H…N, Scheme 1, III) (changes in the surface functional groups were investigated by FT-IR spectroscopy, as shown in Figure S5, with the corresponding details available in the Supporting Information).54 At 12 h, hexagonal prism-like rod structures dominate the precursor (Scheme 1, IV). The hydrothermal reaction facilitates rapid precursor recrystallization at the center of each hexagonal rod, resulting in a higher density of defects than in epitaxial-growth regions.39 Hence, after calcining under N2, pyrolysis-induced collapse occurs from the center of each hexagonal rod in the direction of the tube wall, to form a prism-like hollow tubular g-C3N4. Large amounts of NH3 are released during pyrolysis condensation, which subsequently act as bubble templates that result in abundant pores in the lamellar tube wall (Scheme 1, V).

17 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 51

Scheme 1. Schematic representation of the morphological evolution of CNT-12. Evaluating photocatalytic activity and identifying reactive species The photocatalytic degradation of NO under visible light irradiation was examined in a continuous reactor, the results of which are shown in Figure 3a. NO-adsorption equilibrium in the dark, and reactivity in the absence of a photocatalyst under visible light irradiation were excluded through control experiments. The NO-removal ratios (η) over the tubular g-C3N4 samples are higher than that of bulk g-C3N4. Among them, CNT-12 shows the highest value of η (32.8%) when irradiated by visible light, which is 1.4-, 1.2-, and 1.8-times higher than those of CNT-8 (23.7%), CNT-16 (26.9%), and CN(18.2%), respectively. The Langmuir-Hinshelwood model was used to describe the NO-photodegradation rates because

the

initial

photocatalytic

degradation

of

NO

approximately

follows

mass-transfer-controlled pseudo-first-order kinetics.55-57 (The Langmuir-Hinshelwood model is detailed in the Supporting Information along with calculations of apparent rate constant, k). The apparent reaction rate constant k of CNT-12 (0.045 min−1) was determined to be 2.6, 1.2, 18 ACS Paragon Plus Environment

Page 19 of 51 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

and 2.1 times higher than those of CN (0.017 min−1), CNT-8 (0.037 min−1), and CNT-16 (0.021 min−1), respectively (Figure S6). NO2, as the reaction intermediate, hinders NO degradation because it occupies active sites. Hence, a by-product of NO2 is monitored online (Figure S7a), with selectivity subsequently calculated. The results show that CNT-12 displays the lowest conversion ratio of NO to NO2 among the materials examined, as shown in Figure S7b. After a single-run, the accumulated products on the surfaces of the CN and CNT-12 samples were extracted with deionized water and examined by ion chromatography (IC), which revealed that the amounts of NO3− and NO2− accumulated on the CN surface were 146.7251 μg/g and 30.1789 μg/g, respectively, while those accumulated on the CNT-12 surface were 237.6047 μg/g and 6.5573 μg/g, respectively. In addition, to further examine photocatalytic stability and durability, CNT-12 was subjected to five cycles of use; the NO-removal capability of this material exhibited excellent reproducibility, as shown in Figure 3b. Spin-trap ESR spectroscopy was performed in order to identify the active radicals involved during NO removal over CNT-12. As shown in Figures 3c and b, no ESR signals corresponding to •O2− and •OH were observed in the dark at the commencement of irradiation. However, after 10 min of visible-light irradiation, signals for these two radicals were clearly observed. The observation that the photo-generated holes of CNT-12 are unable to directly oxidize H2O to active •OH species because the EVB of CNT-12 (1.62 V, the band structures of all samples were determined by UV-vis diffuse-reflectance spectroscopy, as detailed in the Supporting Information) is lower than H2O/•OH (2.37 V) indicates that a proportion of •O2− is transformed into •OH radicals via the •O2−→H2O2→•OH route.19 19 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 51

Consequently, signals corresponding to the DMPO-•OH adduct are weaker than those of the DMPO-•O2− adduct. Furthermore, the observation that the EVB of CNT-12 is higher than that of HNO3/NO (0.94 V) implies that holes are able to remove NO by direct oxidation.58 The following holes-scavenger test (The holes-scavenger test is detailed in the Supporting Information) was used to reveal its contribution during photocatalytic NO-removal processes.6 As shown in Figure S8, the value of η of CNT-12 was reduced from 32.8% to 21.3% after introducing KI, indicating that holes were also the active species for NO removal. These results indicate that •O2− and h+ are the dominant active species during NO oxidation over CNT-12, and that •OH radicals make minor contributions due limited pathways for their generation.

Figure 3. (a) Visible light photocatalytic activities of CN, CNT-8, CNT-12, and CNT-16 toward NO removal in air. (b) Cycling of CNT-12 during NO removal. (c) DMPO spin-trap 20 ACS Paragon Plus Environment

Page 21 of 51 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

ESR spectra of CNT-12 dispersed in (a) methanol for DMPO-•O2− and (d) water for DMPO-•OH. Photocatalytic-activity enhancement mechanism According to the ESR results, •O2− radicals are crucial active species involved in the removal of NO. The formation of •O2− involves two steps, namely the adsorption of O2 onto the surface of the material in dark, and the reduction of the adsorbed O2 by photo-generated electrons upon irradiation with light. To investigate the differences in the O2-adsorption behavior of g-C3N4 and N2c-deficient g-C3N4 in detail, O2 adsorption energies and EDDs were calculated by DFT; models of O2 adsorbed on g-C3N4 and N2c-deficient g-C3N4 following geometry optimization are shown in Figures 4a and b, respectively. From the perspective of thermodynamics, molecular O2 is difficult to adsorb onto unbroken g-C3N4 due to the positive ΔEads (0.48 eV). In contrast, it spontaneously adsorbs at the N2c-deficient site because of the negative ΔEads (-5.99 eV). This chemisorption is accompanied by electron transfer between the absorbed O2 and the g-C3N4, with electron coupling further promoting O2 activation. To identify the degree of O2 activation in g-C3N4 and N-deficient g-C3N4, the EDDs of the O2-adsorbed models were calculated, which revealed that about 1.05 electrons are transferred from the N-deficient g-C3N4 to the O2 (Figure 4d), but only 0.05 electrons are transferred from g-C3N4 to O2 (Figure 4c). Furthermore, molecular O2 eventually dissociates into two active O atoms at the N-vacancy site (Figure 4b); these active O atoms are easier to reduce to •O2− than molecular O2 by photo-generated electrons. In addition, models of NO adsorbed on g-C3N4 and N-deficient g-C3N4 are shown in Figures 4e and f, respectively, with corresponding adsorption energies calculated. The calculations reveal that molecular NO is 21 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 51

difficult to adsorb onto unbroken g-C3N4 (Figure 4e, ΔEads = 0.29 eV) but spontaneously adsorbs at the N2c-deficient site (Figure 4f, ΔEads = -5.91 eV). These results indicate that N-vacancies not only increase the O2- and NO-adsorption capacities, but also realize the dissociative activation of molecular O2, which decreases the barrier to the formation of •O2− in subsequent photoreaction processes.

Figure 4. Models of O2 adsorbed on: (a) g-C3N4 and (b) N-deficient g-C3N4 after geometry optimization. The corresponding calculated EDD diagrams of O2-absorbed on: (c) g-C3N4 and (d) N-deficient g-C3N4. Models of NO adsorbed on: (e) g-C3N4 and (f) N-deficient g-C3N4 after geometry optimization. 22 ACS Paragon Plus Environment

Page 23 of 51 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

To illustrate changes in the photoelectric separation efficiency after N-vacancies are introduced into the tubular structure, transient photocurrents and PL spectra were recorded for pristine CN and CNT samples. Figure 5a reveals that the photocurrents generated by the CNT samples are generally higher than that of bulk CN, which suggests that the CNT samples have superior photoinduced charge-separating abilities owing to their hollow tubular structures with thinner layers (Figure S9).59-60 In addition, the CNT samples possess similar hollow tubular structures. The photocurrent was observed to increase from an average value of 0.59 μA/cm2 for CNT-8 to an average value of 1.12 μA/cm2 for CNT-12 owing to the larger BET surface area of the latter, which provides more surface N-vacancies.33,

49

The

photocurrent then decreases to an average value of 0.90 μA/cm2 for CNT-16 owing to the excessive loss of ordered tri-s-triazine framework structures in CNT-16 and its very high concentration of bulk N-defects.33,

49

These results indicated that surface N-vacancies, as

holes traps, are beneficial for reducing photoinduced charge-carrier recombination, leading to more charges and holes that migrate to the catalyst surface and take part in subsequent redox chemistry. However, N-vacancies in the bulk can provide similar pathways, which restrict charges and holes to the bulk and negatively affect the electrochemical response and photocatalytic activity (Figure 5b). Furthermore, as shown in Figure 5c, CNT-12 exhibits the lowest PL intensity and is ~82 % quenched relative to bulk CN, which further suggests that the recombination of photogenerated charges is more substantially suppressed in CNT-12 than in the other samples. To further investigate the differences between the irregular bulk morphology and the tubular structure toward charge-transport capacity, these samples were subjected to EIS in the 23 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 51

dark. Figure 5d reveals that the radii of the arcs in the Nyquist plots of the CNT samples are smaller than that of CN, indicating that 1D tubular structures facilitate facile charge-carrier transport in their longitudinal directions.40, 61 Among the tubular structures, CNT-12, with the smallest arc radius, exhibited the highest charge-transfer efficiency, owing to its suitable thin-wall construction and longitudinal structure; however the performance of CNT-16, with a similar thin-wall structure, was unsatisfactory, which is ascribable to the loss of intralayer long-range ordered tri-s-triazine units (as evidenced by XRD, see Figure 1a). These above-mentioned results demonstrate that suitable surface N-vacancies and 1D tubular ordered structures are responsible for effective electron-hole separation and transfer, leading to enhanced photocatalytic activity.

Figure 5. (a) Transient photocurrent responses of CN, CNT-8, CNT-12, and CNT-16. (b) Schematic diagram depicting the roles of N-vacancies in the migration and separation 24 ACS Paragon Plus Environment

Page 25 of 51 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

behavior of photoinduced charges and holes pairs, and (c) PL spectra and (d) Nyquist plots of CN, CNT-8, CNT-12, and CNT-16. Moreover, the optical-absorbing abilities and bandgaps of the prepared g-C3N4 samples were examined by UV-vis diffuse reflectance spectroscopy, the results of which are shown in Figure 6a. Compared to bulk CN, clear blue-shifts in the intrinsic absorption edges, from 458 nm for CN to 451, 446, and 447 nm for CNT-8, CNT-12, and CNT-16 were observed, respectively, which are the result of increasingly thin layers that reduce interlayer electron coupling.60 However, increased absorption tails are observed in the visible-light regions of the spectra of the CNT samples that are due to N-defect states.62 The electronic structures of g-C3N4 and N-deficient g-C3N4 were calculated based on the atomic-structure models shown in Figure S1. Compared with the electronic structure of g-C3N4 (Figure S10), a defect energy level is observed for N-deficient g-C3N4, as illustrated in Figure 6a, which coincides with the increasing absorption tail observed for the CNT samples (Figure 6a). The bandgaps were determined by the Kubelka–Munk method to be 2.71, 2.75, 2.78, and 2.77 eV for CN, CNT-8, CNT-12, and CNT-16, respectively (Figure 6b). The valence-band (VB) and conduction-band (CB) potentials of a semiconductor material can be determined using empirical equations, the details of which are provided in the Supporting Information; the positions of the VBs and CBs of the as-prepared samples were calculated to be 1.59 and -1.13 V for CN, 1.60 and -1.15 V for CNT-8, 1.62 and -1.16 V for CNT-12, and 1.62 and -1.15 V for CNT-16, respectively. Hence, N-vacancies simultaneously enhance the visible-light absorbing ability of g-C3N4, while slightly enhancing its redox capabilities.

25 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 6. (a) UV-vis absorption spectra of CN, CNT-8, CNT-12, and CNT-16, and the calculated electronic structure (inset) of N-deficient g-C3N4. (b) Transformed Kubelka–Munk functions of light-absorbed CN, CNT-8, CNT-12, and CNT-16. Adsorption sites, and the identification of intermediates and photocatalytic reaction pathways Based on the aforementioned DFT results for O2 and NO adsorption, N-vacancies are proposed to be the active sites that adsorption-activate reactants. In order to further verify this hypothesis, the NO- and O2-adsorption processes over CN and CNT-12 were investigated by in-situ DRIFTS, the results of which are displayed in Figures 7a and b, respectively. A background spectrum was recorded after pretreating each sample with high-purity He; detailed descriptions of the NO- and O2-adsorption processes over CN are provided in the Supporting Information, which revealed that that both NO and O2 adsorb at the N–H bonds of CN to form NOH and OOH bonds; NO– and surface peroxo species are formed as a result of these activation processes. Compared to the adsorption reactions of CN (Figure 7a), some clear differences are observed when NO and O2 are adsorbed by CNT-12. Firstly, two new adsorption peaks were observed to gradually appear at 1350 and 1299 cm-1 in the spectra shown in Figure 7b; these peaks are assigned to adsorbed nitro compounds (–NO2).63-64 A 26 ACS Paragon Plus Environment

Page 26 of 51

Page 27 of 51 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

DFT model of NO and O2 co-adsorbed on N-deficient g-C3N4 was built and is displayed in Figure S11; this model reveals that one of the active O atoms of O2 binds to the N atom of NO to form the –NO2 group following geometry optimization. In addition, a bidentate-state stretching band at 1024 cm-1 was observed.63-64 These results indicate that N-vacancies can indeed absorb and activate NO and O2 molecules. In addition, the NO2 (3604–3740 cm-1), NO– (1095 cm-1), N2O4 (920 cm-1), and surface peroxo (860 and 786 cm-1) peaks in the spectrum of the CNT-12 surface are more obvious than those in the spectrum of the CN surface. However, the intensity of the NOH peak (3561 cm-1) on the CNT-12 surface is weaker than that on the CN surface. These results indicate that NO and O2 are more rapidly absorbed at N-vacancies through bidentate states than terminal N-H bonds due to extra active electrons. Once adsorption equilibrium is established, a visible-light source was applied to initiate photoreaction processes. Time-dependent DRIFTS spectra for the photocatalytic oxidation of NO over CN and CNT-12 are displayed Figures 7c and d, respectively, with detailed descriptions of the photo-reactions involving CN provided in the Supporting Information. Compared to the characteristic peaks of CN during photo-reaction processes, new absorption bands were observed at 1500–1600 (monodentate nitrate, ν(NO3)), 1226 (bidentate nitrate, ν(NO3)), and 1192 (bidentate nitrite, ν(NO2)) cm-1.64 In addition, except for a slight decrease in the intensity of the 3562 cm-1 peak, peaks corresponding to other types of nitrates and peroxo species on the CNT-12 surface were more intense than those on the CN surface. These results demonstrate that the CNT-12 surface is more active than that of CN during both adsorption and photocatalytic-reaction phases. Notably, these new peaks almost disappear 27 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 51

when the light was switched off after 20 min, with the adsorption equilibrium re-established in both CN and CNT-12, consistent with NO3– dissociation. Based on the aforementioned chain of evidence, we conclude that different reaction pathways and reactive sites are involved in the adsorption and photocatalytic oxidation of NO over pristine and N-deficient g-C3N4, as depicted in Figure 7e. Simply, NO and O2 are absorbed at N–H bonds to form NOH and OOH bonds on pristine g-C3N4, (Figure 7e, left (I)), which are then activated to form NO– and surface peroxo species (Figure 7e, left (II)) during adsorption. Upon illumination, peroxo species are reduced by photogenerated electrons to form •O2− (Figure 7e, left (III)) that, together with photogenerated h+, oxidize absorbed-state intermediates to afford the end product (NO3-) (Figure 7e, left (IV)). However, NO and O2 adsorption and photocatalysis over N-deficient g-C3N4 proceed via a different pathway. NO and O2 are more rapidly absorbed at N-vacancies (Figure 7e, right (I)) through bidentate states, rather than at terminal N-H bonds; these species are activated and form nitro compounds (–NO2) (Figure 7e, right (II)) rather than NOH or OOH bonds. Upon irradiation, the absorbed-state intermediates are oxidized to NO3– by •O2− and h+ at N-vacancies, as shown in Figure 7e, right (III).

28 ACS Paragon Plus Environment

Page 29 of 51 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Figure 7. Time-dependent DRIFTS spectra for the adsorption of NO and O2, and their photocatalytic reactions on (a, c) CN and (b, d) CNT-12, respectively. (e) Depicting the proposed reaction pathways for adsorption and the photocatalytic oxidation of NO over pristine (left) and N-deficient (right) g-C3N4.

CONCLUSION In summary, in this study we demonstrated a feasible method for the preparation of 29 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 30 of 51

N-deficient g-C3N4 with a porous tubular micro-nanostructure that enhances O2 activation, photoinduced charge-carrier separation, and oriented transfer. The optimal sample, CNT-12, exhibited a NO removal ratio under visible-light irradiation that was 1.8 times higher than that of bulk g-C3N4. The spatial distribution of N-vacancies activates and dissociates molecular O2 while also affecting electron-separation behavior. The porous tubular architecture favors the oriented transfer of charge carriers and the diffusion of reactants. These factors contribute to the observed enhancement in photocatalytic activity compared to the

bulk

material.

In-situ

DRIFTS

revealed

that

NO

and

O2

participate

in

adsorption-/photo-reactions at N-vacancies in N-deficient g-C3N4, and at terminal N-H bonds in pristine g-C3N4. This study has established a feasible strategy for modifying g-C3N4 through simultaneous N-vacancy incorporation and morphology modulation, which improves photocatalytic activity toward NOx degradation.

ASSOCIATED CONTENT Supporting Information. Structure models of g-C3N4 and N-deficient g-C3N4; Element contents for M, M-8, M-12, M-16, CN, CNT-8, CNT-12 and CNT-16 samples; Surface N/C atomic ratio, N2c/N3c area ratio of obtained g-C3N4 samples determined by XPS and bulk N/C atomic ratio of obtained precursors and g-C3N4 samples by elemental analyzer; XPS analysis of CNT-8, CNT-12 and CNT-16 samples; XPS analysis of oxygen concentration of CN, CNT-8, CNT-12 and CNT-16 samples; High-resolution O 1s XPS of CN, CNT-8, CNT-12, and CNT-16; SEM images of CN; TEM images of CNT-8 and CNT-16; SEM images of M, M-6 and M-7; XRD patterns of single melamine treated by different hydrothermal times (6, 30 ACS Paragon Plus Environment

Page 31 of 51 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

7, 8, 12, 16 h); Schematic crystal structure of melamine and melamine-cyanuric acid supramolecular precursor; FT-IR spectra of single melamine treated by different hydrothermal times: 0, 6, 7, 8, 12, 16 h and CA; Reaction rate constants k of CN, CNT-8, CNT-12 and CNT-16; The monitoring of the fraction of NO2 in the outlet of the reactor and NO2 selectivity of the CN, CNT-8, CNT-12 and CNT-16, respectively; Visible light photocatalytic activities of CNT-12 toward NO removal in air with/without the addition of KI; Schematic diagram of photoinduced carriers separation processes in bulk and layered structure of g-C3N4; The calculated electronic structures of g-C3N4; A DFT model of NO and O2 co-adsorption on N-deficient g-C3N4.

AUTHOR INFORMATION Corresponding Author * Prof. Yu Huang, E-mail address: [email protected]

Tel: 86-29-6233 6261

Author Contributions The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. Notes The authors declare no competing financial interest.

ACKNOWLEDGMENT This research was financially supported by the National Key Research and Development 31 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 32 of 51

Program of China (2016YFA0203000 and 2017YFC0212200) and the National Science Foundation of China (No. 41573138 and No. 51878644). Moreover, we acknowledge the partial support by the Key Research and Development Program of Shaanxi Province (No. 2018ZDCXL-SF-02-04). Yu Huang was also supported by the “Hundred Talent Program” of the Chinese Academy of Sciences.

REFERENCES (1) Gong, X.; Kaulfus, A.; Nair, U.; Jaffe, D. A. Quantifying O3 Impacts in Urban Areas Due to Wildfires Using a Generalized Additive Model. Environ. Sci. Technol. 2017, 51 (22), 13216-13223. (2) Huang, R.-J.; Zhang, Y.; Bozzetti, C.; Ho, K.-F.; Cao, J.-J.; Han, Y.; Daellenbach, K. R.; Slowik, J. G.; Platt, S. M.; Canonaco, F.; Zotter, P.; Wolf, R.; Pieber, S. M.; Bruns, E. A.; Crippa, M.; Ciarelli, G.; Piazzalunga, A.; Schwikowski, M.; Abbaszade, G.; Schnelle-Kreis, J.; Zimmermann, R.; An, Z.; Szidat, S.; Baltensperger, U.; El Haddad, I.; Prevot, A. S. H. High Secondary Aerosol Contribution to Particulate Pollution during Haze Events in China. Nature 2014, 514 (7521), 218-222. (3) Chen, H.; Nanayakkara, C. E.; Grassian, V. H. Titanium Dioxide Photocatalysis in Atmospheric Chemistry. Chem. Rev. 2012, 112 (11), 5919-5948. (4) Huang, Y.; Liang, Y.; Rao, Y.; Zhu, D.; Cao, J.-j.; Shen, Z.; Ho, W.; Lee, S. C. Environment-Friendly Carbon Quantum Dots/ZnFe2O4 Photocatalysts: Characterization, Biocompatibility, and Mechanisms for NO Removal. Environ. Sci. Technol. 2017, 51 (5), 2924-2933. 32 ACS Paragon Plus Environment

Page 33 of 51 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

(5) Wang, Z.; Yan, S.; Sun, Y.; Xiong, T.; Dong, F.; Zhang, W. Bi Metal Sphere/Graphene Oxide Nanohybrids with Enhanced Direct Plasmonic Photocatalysis. Appl. Catal. B: Environ. 2017, 214, 148-157. (6) Zhang, Q.; Huang, Y.; Peng, S.; Zhang, Y.; Shen, Z.; Cao, J.-j.; Ho, W.; Lee, S. C.; Pui, D. Y. H. Perovskite LaFeO3-SrTiO3 Composite for Synergistically Enhanced NO Removal Under Visible Light Excitation. Appl. Catal. B: Environ. 2017, 204, 346-357. (7) Fujiwara, K.; Mueller, U.; Pratsinis, S. E. Pd Subnano-Clusters on TiO2 for Solar-Light Removal of NO. ACS Catal. 2016, 6 (3), 1887-1893. (8) Wang, W.; Huang, Y.; Zhang, X.; Cao, J. J.; Ho, W.; Lee, S. C. Three-Dimensional Bi5O7I Photocatalysts for Efficient Removal of NO in Air Under Visible Light. Aerosol Sci. Eng. 2017, 1 (1), 33-40. (9) Jia, Y.; Li, S.; Gao, J.; Zhu, G.; Zhang, F.; Shi, X.; Huang, Y.; Liu, C. Highly Efficient (BiO)2CO3-BiO2-x-graphene Photocatalysts: Z-Scheme Photocatalytic Mechanism for Their Enhanced Photocatalytic Removal of NO. Appl. Catal. B: Environ. 2019, 240, 241-252. (10) Choi, C. H.; Lin, L.; Gim, S.; Lee, S.; Kim, H.; Wang, X.; Choi, W. Polymeric Carbon Nitride with Localized Aluminum Coordination Sites as a Durable and Efficient Photocatalyst for Visible Light Utilization. ACS Catal. 2018, 8 (5), 4241-4256. (11) Ong, W.-J.; Tan, L.-L.; Ng, Y. H.; Yong, S.-T.; Chai, S.-P. Graphitic Carbon Nitride (g-C3N4)-Based Photocatalysts for Artificial Photosynthesis and Environmental Remediation: Are We a Step Closer To Achieving Sustainability? Chem. Rev. 2016, 116 (12), 7159-7329. (12) Dong, G.; Yang, L.; Wang, F.; Zang, L.; Wang, C. Removal of Nitric Oxide through Visible Light Photocatalysis by g-C3N4 Modified with Perylene Imides. ACS Catal. 2016, 6 33 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 34 of 51

(10), 6511-6519. (13) Wang, X.; Maeda, K.; Thomas, A.; Takanabe, K.; Xin, G.; Carlsson, J. M.; Domen, K.; Antonietti, M. A Metal-free Polymeric Photocatalyst for Hydrogen Production from Water under Visible Light. Nat. Mater. 2009, 8 (1), 76-80. (14) Chen, P.; Xing, P.; Chen, Z.; Lin, H.; He, Y. Rapid and Energy-efficient Preparation of Boron Doped g-C3N4 with Excellent Performance in Photocatalytic H2-evolution. Int. J. Hydrogen. Energ. 2018, 43 (43), 19984-19989. (15) He, Y.; Zhang, L.; Teng, B.; Fan, M. New Application of Z-scheme Ag3PO4/g-C3N4 Composite in Converting CO2 to Fuel. Environ. Sci. Technol. 2014, 49 (1), 649-656. (16) Shi, L.; Liang, L.; Wang, F.; Liu, M.; Chen, K.; Sun, K.; Zhang, N.; Sun, J. Higher Yield Urea-derived Polymeric Graphitic Carbon Nitride with Mesoporous Structure and Superior Visible-light-responsive Activity. ACS Sustain. Chem. Eng. 2015, 3 (12), 3412-3419. (17) Zou, Y.; Shi, J.-W.; Ma, D.; Fan, Z.; Cheng, L.; Sun, D.; Wang, Z.; Niu, C. WS2/Graphitic Carbon Nitride Heterojunction Nanosheets Decorated with CdS Quantum Dots for Photocatalytic Hydrogen Production. ChemSusChem 2018, 11 (7), 1187-1197. (18) Wang, Z.; Huang, Y.; Chen, L.; Chen, M.; Cao, J.; Ho, W.; Lee, S. C. In situ g-C3N4 Self-sacrificial Synthesis of a g-C3N4/LaCO3OH Heterostructure with Strong Interfacial Charge Transfer and Separation for Photocatalytic NO Removal. J. Mater. Chem. A 2018, 6 (3), 972-981. (19) Cui, W.; Li, J.; Dong, F.; Sun, Y.; Jiang, G.; Cen, W.; Lee, S. C.; Wu, Z. Highly Efficient Performance and Conversion Pathway of Photocatalytic NO Oxidation on SrO-Clusters@Amorphous Carbon Nitride. Environ. Sci. Technol. 2017, 51 (18), 34 ACS Paragon Plus Environment

Page 35 of 51 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

10682-10690. (20) Zhu, J.; Carabineiro, S. A. C.; Shan, D.; Faria, J. L.; Zhu, Y.; Figueiredo, J. L. Oxygen Activation Sites in Gold and Iron Catalysts Supported on Carbon Nitride and Activated Carbon. J. Catal. 2010, 274 (2), 207-214. (21) Hong, Z.; Shen, B.; Chen, Y.; Lin, B.; Gao, B. Enhancement of Photocatalytic H2 Evolution over Nitrogen-deficient Graphitic Carbon Nitride. J. Mater. Chem. A 2013, 1 (38), 11754-11761. (22) Zou, Y.; Shi, J.-W.; Ma, D.; Fan, Z.; He, C.; Cheng, L.; Sun, D.; Li, J.; Wang, Z.; Niu, C. Efficient Spatial Charge Separation and Transfer in Ultrathin g-C3N4 Nanosheets Modified with Cu2MoS4 as a Noble Metal-free Co-catalyst for Superior Visible Light-driven Photocatalytic Water Splitting. Catal. Sci. Technol. 2018, 8 (15), 3883-3893. (23) Zou, Y.; Ma, D.; Sun, D.; Mao, S.; He, C.; Wang, Z.; Ji, X.; Shi, J.-W. Carbon Nanosheet Facilitated Charge Separation and Transfer between Molybdenum Carbide and Graphitic Carbon Nitride toward Efficient Photocatalytic H2 Rroduction. Appl. Surf. Sci. 2019, 473, 91-101. (24) Liu, Y.; Xiao, C.; Li, Z.; Xie, Y. Vacancy Engineering for Tuning Electron and Phonon Structures of Two-Dimensional Materials. Adv. Energy Mater. 2016, 6 (23): 1600436. (25) Gong, X.-Q.; Selloni, A.; Batzill, M.; Diebold, U. Steps on Anatase TiO2(101). Nat. Mater. 2006, 5 (8), 665-670. (26) Liu, G.; Yang, H. G.; Wang, X.; Cheng, L.; Lu, H.; Wang, L.; Lu, G. Q.; Cheng, H.-M. Enhanced Photoactivity of Oxygen-Deficient Anatase TiO2 Sheets with Dominant {001} Facets. J. Phys. Chem. C 2009, 113 (52), 21784-21788. 35 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 36 of 51

(27) Liu, G.; Yin, L.-C.; Wang, J.; Niu, P.; Zhen, C.; Xie, Y.; Cheng, H.-M. A Red Anatase TiO2 Photocatalyst for Solar Energy Conversion. Energy Environ. Sci. 2012, 5 (11), 9603-9610. (28) Chen, G.; Zhao, Y.; Fu, G.; Duchesne, P. N.; Gu, L.; Zheng, Y.; Weng, X.; Chen, M.; Zhang, P.; Pao, C.-W.; Lee, J.-F.; Zheng, N. Interfacial Effects in Iron-Nickel Hydroxide-Platinum Nanoparticles Enhance Catalytic Oxidation. Science 2014, 344 (6183), 495-499. (29) Lu, Y.; Huang, Y.; Zhang, Y.; Cao, J.-j.; Li, H.; Bian, C.; Lee, S. C. Oxygen Vacancy Engineering of Bi2O3/Bi2O2CO3 Heterojunctions: Implications of the Interfacial Charge Transfer, NO Adsorption and Removal. Appl. Catal. B: Environ. 2018, 231, 357-367. (30) Tay, Q.; Kanhere, P.; Ng, C. F.; Chen, S.; Chakraborty, S.; Huan, A. C. H.; Sum, T. C.; Ahuja, R.; Chen, Z. Defect Engineered g-C3N4 for Efficient Visible Light Photocatalytic Hydrogen Production. Chem. Mater. 2015, 27 (14), 4930-4933. (31) Wu, P.; Wang, J.; Zhao, J.; Guo, L.; Osterloh, F. E. Structure Defects in g-C3N4 Limit Visible Light Driven Hydrogen Evolution and Photovoltage. J. Mater. Chem. A 2014, 2 (47), 20338-20344. (32) Niu, P.; Liu, G.; Cheng, H.-M. Nitrogen Vacancy-Promoted Photocatalytic Activity of Graphitic Carbon Nitride. J. Phys. Chem. C 2012, 116 (20), 11013-11018. (33) Yu, H.; Shi, R.; Zhao, Y.; Bian, T.; Zhao, Y.; Zhou, C.; Waterhouse, G. I. N.; Wu, L.-Z.; Tung, C.-H.; Zhang, T. Alkali-Assisted Synthesis of Nitrogen Deficient Graphitic Carbon Nitride with Tunable Band Structures for Efficient Visible-Light-Driven Hydrogen Evolution. Adv. Mater. 2017, 29 (16): 1605148. 36 ACS Paragon Plus Environment

Page 37 of 51 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

(34) Wu, N.; Wang, J.; Tafen, D. N.; Wang, H.; Zheng, J.-G.; Lewis, J. P.; Liu, X.; Leonard, S. S.; Manivannan, A. Shape-Enhanced Photocatalytic Activity of Single-Crystalline Anatase TiO2 (101) Nanobelts. J. Am. Chem. Soc. 2010, 132 (19), 6679-6685. (35) Mor, G. K.; Shankar, K.; Paulose, M.; Varghese, O. K.; Grimes, C. A. Use of Highly-ordered TiO2 Nanotube Arrays in Dye-sensitized Solar Cells. Nano Lett. 2006, 6 (2), 215-218. (36) Jun, Y.-S.; Park, J.; Lee, S. U.; Thomas, A.; Hong, W. H.; Stucky, G. D. Three-Dimensional Macroscopic Assemblies of Low-Dimensional Carbon Nitrides for Enhanced Hydrogen Evolution. Angew. Chem. Int. Edit. 2013, 52 (42), 11083-11087. (37) Shalom, M.; Inal, S.; Fettkenhauer, C.; Neher, D.; Antonietti, M. Improving Carbon Nitride Photocatalysis by Supramolecular Preorganization of Monomers. J. Am. Chem. Soc. 2013, 135 (19), 7118-7121. (38) Wang, Z.; Chen, M.; Huang, Y.; Shi, X.; Zhang, Y.; Huang, T.; Cao, J.; Ho, W.; Lee, S. C. Self-assembly Synthesis of Boron-doped Graphitic Carbon Nitride Hollow Tubes for Enhanced Photocatalytic NOx Removal Under Visible Light. Appl. Catal. B: Environ. 2018, 239, 352-361. (39) Guo, S.; Deng, Z.; Li, M.; Jiang, B.; Tian, C.; Pan, Q.; Fu, H. Phosphorus-Doped Carbon Nitride Tubes with a Layered Micro-nanostructure for Enhanced Visible-Light Photocatalytic Hydrogen Evolution. Angew. Chem. Int. Edit. 2016, 55 (5), 1830-1834. (40) Tong, Z.; Yang, D.; Sun, Y.; Nan, Y.; Jiang, Z. Tubular g-C3N4 Isotype Heterojunction: Enhanced Visible-Light Photocatalytic Activity through Cooperative Manipulation of Oriented Electron and Hole Transfer. Small 2016, 12 (30), 4093-4101. 37 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 38 of 51

(41) Wang, Z.; Guan, W.; Sun, Y.; Dong, F.; Zhou, Y.; Ho, W.-K. Water-assisted Production of Honeycomb-like g-C3N4 with ultralong carrier lifetime and outstanding photocatalytic activity. Nanoscale 2015, 7 (6), 2471-2479. (42) Kresse, G.; Furthmüller, J. Efficient Iterative Schemes for ab Initio Total-energy Calculations Using a Plane-wave Basis Set. Phys. Rev. B 1996, 54 (16):11169. (43) Kresse, G.; Joubert, D. From Ultrasoft Pseudopotentials to the Projector Augmented-wave Method. Phys. Rev. B 1999, 59 (3): 1758. (44) Perdew, J. P.; Burke, K.; Ernzerhof, M. Generalized Gradient Approximation Made Simple. Phys. Rev. Lett. 1996, 77 (18), 3865-3868. (45) Blochl, P. E. Projector Augmented-wave Method. Phys. Rev. B 1994, 50 (24), 17953-17979. (46) Mulliken, R. S. Electronic Population Analysis on Lcao-mo Molecular Wave Functions .1. J. Chem. Phys. 1955, 23 (10), 1833-1840. (47) Chen, P.; Xing, P.; Chen, Z.; Hu, X.; Lin, H.; Zhao, L.; He, Y. In-situ Synthesis of AgNbO3/g-C3N4 Photocatalyst via Microwave Heating Method for Efficiently Photocatalytic H2 Generation. J. Colloid Interf. Sci 2019, 534, 163-171. (48) Wang, Z.; Huang, Y.; Ho, W.; Cao, J.; Shen, Z.; Lee, S. C. Fabrication of Bi2O2CO3/g-C3N4 Heterojunctions for Efficiently Photocatalytic NO in Air Removal: In-situ Self-sacrificial Synthesis, Characterizations and Mechanistic Study. Appl. Catal. B: Environ. 2016, 199, 123-133. (49) Niu, P.; Yin, L.-C.; Yang, Y.-Q.; Liu, G.; Cheng, H.-M. Increasing the Visible Light Absorption of Graphitic Carbon Nitride (Melon) Photocatalysts by Homogeneous 38 ACS Paragon Plus Environment

Page 39 of 51 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Self-Modification with Nitrogen Vacancies. Adv. Mater. 2014, 26 (47), 8046-8052. (50) Shi, J.-W.; Zou, Y.; Ma, D.; Fan, Z.; Cheng, L.; Sun, D.; Wang, Z.; Niu, C.; Wang, L. Stable 1T-phase MoS2 as an Effective Electron Mediator Promoting Photocatalytic Hydrogen Production. Nanoscale 2018, 10 (19), 9292-9303. (51) Shi, L.; Liang, L.; Wang, F.; Liu, M.; Sun, J. Ag2CrO4 Nanoparticles Loaded on Two-dimensional Large Surface Area Graphite-like Carbon Nitride Sheets: Simple Synthesis and Excellent Photocatalytic Performance. Dalton. Trans. 2016, 45 (13), 5815-5824. (52) Fu, J.; Zhu, B.; Jiang, C.; Cheng, B.; You, W.; Yu, J. Hierarchical Porous O-Doped g-C3N4 with Enhanced Photocatalytic CO2 Reduction Activity. Small 2017, 13 (15): 1603938. (53) Shen, J.-S.; Cai, Q.-G.; Jiang, Y.-B.; Zhang, H.-W. Anion-triggered Melamine Based Self-assembly and Hydrogel. Chem. Commun. 2010, 46 (36), 6786-6788. (54) Jun, Y.-S.; Lee, E. Z.; Wang, X.; Hong, W. H.; Stucky, G. D.; Thomas, A. From Melamine-Cyanuric Acid Supramolecular Aggregates to Carbon Nitride Hollow Spheres. Adv. Funct. Mater. 2013, 23 (29), 3661-3667. (55) Shie, J.-L.; Lee, C.-H.; Chiou, C.-S.; Chang, C.-T.; Chang, C.-C.; Chang, C.-Y. Photodegradation Kinetics of Formaldehyde Using light sources of UVA, UVC and UVLED in the Presence of Composed Silver Titanium Oxide Photocatalyst. J. Hazard. Mater. 2008, 155 (1-2), 164-172. (56) Ai, Z.; Ho, W.; Lee, S.; Zhang, L. Efficient Photocatalytic Removal of NO in Indoor Air with Hierarchical Bismuth Oxybromide Nanoplate Microspheres under Visible Light. Environ. Sci. Technol. 2009, 43 (11), 4143-4150. 39 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 40 of 51

(57) Chen, M.; Yao, J.; Huang, Y.; Gong, H.; Chu, W. Enhanced Photocatalytic Degradation of Ciprofloxacin over Bi2O3/(BiO)2CO3 Heterojunctions: Efficiency, Kinetics, Pathways, Mechanisms and Toxicity Evaluation. Chem. Eng. J. 2018, 334, 453-461. (58) Ge, L.; Han, C.; Liu, J. Novel Visible Light-induced g-C3N4/Bi2WO6 composite photocatalysts for efficient degradation of methyl orange. Appl. Catal. B: Environ. 2011, 108 (1-2), 100-107. (59) Liu, G.; Wang, T.; Zhang, H.; Meng, X.; Hao, D.; Chang, K.; Li, P.; Kako, T.; Ye, J. Nature-Inspired Environmental "Phosphorylation" Boosts Photocatalytic H2 Production over Carbon Nitride Nanosheets under Visible-Light Irradiation. Angew. Chem. Int. Edit. 2015, 54 (46), 13561-13565. (60) Niu, P.; Zhang, L.; Liu, G.; Cheng, H.-M. Graphene-Like Carbon Nitride Nanosheets for Improved Photocatalytic Activities. Adv. Funct. Mater. 2012, 22 (22), 4763-4770. (61) Ran, J.; Ma, T. Y.; Gao, G.; Du, X.-W.; Qiao, S. Z. Porous P-doped Graphitic Carbon Nitride Nanosheets for Synergistically Enhanced Visible-light Photocatalytic H2 Production. Energy Environ. Sci. 2015, 8 (12), 3708-3717. (62) Lu, L.; Wang, B.; Wang, S.; Shi, Z.; Yan, S.; Zou, Z. La2O3-Modified LaTiO2N Photocatalyst with Spatially Separated Active Sites Achieving Enhanced CO2 Reduction. Adv. Funct. Mater. 2017, 27 (36): 1702447. (63) Kantcheva, M. Identification, Stability, and Reactivity of NOx Species Adsorbed on Titania-supported Manganese Catalysts. J. Catal. 2001, 204 (2), 479-494. (64) Wu, J. C. S.; Cheng, Y. T. In situ FTIR study of Photocatalytic NO Reaction on Photocatalysts under UV Irradiation. J. Catal. 2006, 237 (2), 393-404. 40 ACS Paragon Plus Environment

Page 41 of 51 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Figure Captions Figure 1. (a) XRD patterns, and (b) FT-IR, (c) high-resolution C 1s and N 1s XPS, and (e) ESR spectra of CN, CNT-8, CNT-12, and CNT-16. (d) Calculated EDDs diagrams of g-C3N4 (left) and N2c-deficient g-C3N4 (right). Figure 2. SEM images of (a) CNT-8, (b) CNT-12, and (c) CNT-16. (d) TEM image of CNT-12. (e) Enlarged TEM image of the tubular wall of CNT-12. (f) N2 adsorption-desorption isotherms of CN, CNT-8, CNT-12, and CNT-16, and the corresponding pore-size distribution curves. Figure 3. (a) Visible light photocatalytic activities of CN, CNT-8, CNT-12, and CNT-16 toward NO removal in air. (b) Cycling of CNT-12 during NO removal. (c) DMPO spin-trap ESR spectra of CNT-12 dispersed in (a) methanol for DMPO-•O2− and (d) water for DMPO-•OH. Figure 4. Models of O2 adsorbed on: (a) g-C3N4 and (b) N-deficient g-C3N4 after geometry optimization. The corresponding calculated EDD diagrams of O2-absorbed on: (c) g-C3N4 and (d) N-deficient g-C3N4. Models of NO adsorbed on: (e) g-C3N4 and (f) N-deficient g-C3N4 after geometry optimization. Figure 5. (a) Transient photocurrent responses of CN, CNT-8, CNT-12, and CNT-16. (b) Schematic diagram depicting the roles of N-vacancies in the migration and separation behavior of photoinduced charges and holes pairs, and (c) PL spectra and (d) Nyquist plots of CN, CNT-8, CNT-12, and CNT-16. 41 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 42 of 51

Figure 6. (a) UV-vis absorption spectra of CN, CNT-8, CNT-12, and CNT-16, and the calculated electronic structure (inset) of N-deficient g-C3N4. (b) Transformed Kubelka–Munk functions of light-absorbed CN, CNT-8, CNT-12, and CNT-16. Figure 7. Time-dependent DRIFTS spectra for the adsorption of NO and O2, and their photocatalytic reactions on (a, c) CN and (b, d) CNT-12, respectively. (e) Depicting the proposed reaction pathways for adsorption and the photocatalytic oxidation of NO over pristine (left) and N-deficient (right) g-C3N4. Scheme 1. Schematic representation of the morphological evolution of CNT-12.

42 ACS Paragon Plus Environment

Page 43 of 51 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Figure 1. (a) XRD patterns, and (b) FT-IR, (c) high-resolution C 1s and N 1s XPS, and (e) ESR spectra of CN, CNT-8, CNT-12, and CNT-16. (d) Calculated EDDs diagrams of g-C3N4 (left) and N2c-deficient g-C3N4 (right).

43 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 44 of 51

Figure 2. SEM images of (a) CNT-8, (b) CNT-12, and (c) CNT-16. (d) TEM image of CNT-12. (e) Enlarged TEM image of the tubular wall of CNT-12. (f) N2 adsorption-desorption isotherms of CN, CNT-8, CNT-12, and CNT-16, and the corresponding pore-size distribution curves. 44 ACS Paragon Plus Environment

Page 45 of 51 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Figure 3. (a) Visible light photocatalytic activities of CN, CNT-8, CNT-12, and CNT-16 toward NO removal in air. (b) Cycling of CNT-12 during NO removal. (c) DMPO spin-trap ESR spectra of CNT-12 dispersed in (a) methanol for DMPO-•O2− and (d) water for DMPO-•OH.

45 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 46 of 51

Figure 4. Models of O2 adsorbed on: (a) g-C3N4 and (b) N-deficient g-C3N4 after geometry optimization. The corresponding calculated EDD diagrams of O2-absorbed on: (c) g-C3N4 and (d) N-deficient g-C3N4. Models of NO adsorbed on: (e) g-C3N4 and (f) N-deficient g-C3N4 after geometry optimization.

46 ACS Paragon Plus Environment

Page 47 of 51 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Figure 5. (a) Transient photocurrent responses of CN, CNT-8, CNT-12, and CNT-16. (b) Schematic diagram depicting the roles of N-vacancies in the migration and separation behavior of photoinduced charges and holes pairs, and (c) PL spectra and (d) Nyquist plots of CN, CNT-8, CNT-12, and CNT-16.

47 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 48 of 51

Figure 6. (a) UV-vis absorption spectra of CN, CNT-8, CNT-12, and CNT-16, and the calculated electronic structure (inset) of N-deficient g-C3N4. (b) Transformed Kubelka–Munk functions of light-absorbed CN, CNT-8, CNT-12, and CNT-16.

48 ACS Paragon Plus Environment

Page 49 of 51 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Figure 7. Time-dependent DRIFTS spectra for the adsorption of NO and O2, and their photocatalytic reactions on (a, c) CN and (b, d) CNT-12, respectively. (e) Depicting the proposed reaction pathways for adsorption and the photocatalytic oxidation of NO over pristine (left) and N-deficient (right) g-C3N4.

49 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 50 of 51

Scheme 1. Schematic representation of the morphological evolution of CNT-12.

50 ACS Paragon Plus Environment

Page 51 of 51 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

For Table of Contents Only 83x35mm (300 x 300 DPI)

ACS Paragon Plus Environment